Extrapolation simulation for estimating human avoidability in human-robot coexistence systems
Koji Sunada, Yoji Yamada, Takamasa Hattori, Shogo Okamoto, Susumu Hara
- 发表年份
- 2012
- 引用次数
- 6
摘要
In the design of human-robot coexistence environments, human avoidance actions are not taken into consideration. We conduct a psychological experiment to observe human avoidance actions where a mechanical harm approached his or her eyes. Using the observed motions, we carry out “extrapolation simulation” of human avoidance actions in collision situations. As simulation results, first, we express collision probability using the motion range of the end effector of a robot as a variable. Second, we identify the motion conditions, which ensure the safety, of the end effector in our setup. Finally, we quantitatively compare the safety conditions between taking and not taking avoidance actions into consideration. This study contributes to providing a tool for conducting a practical risk assessment aimed at realizing a safe human-robot coexistence system.
关键词
相关论文
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991
A new optimizer using particle swarm theory
R.C. Eberhart, James Kennedy
2002